1. Home
  2. Electronics Design
  3. charging the bootstrap cap

charging the bootstrap cap

Lots of half-bridge and full-bridge mosfet gate drivers use a bootstrap capacitor to make the necessary gate drive for the upper n-fet. The upper-fet gate needs to swing maybe 10 volts above the drain supply, which is generally the biggest voltage around.

But the first time the driver is enabled, the boot cap may not be charged. And it will eventually discharge, given a long enough positive output pulse. I've never seen a data sheet mention this.

I think I recall some driver that has a charge pump, but that's rare.

Halfway driving a fet is a good way to blow it up. I haven't done that this week, but it is a concern.

Reply to
John Larkin
Loading thread data ...

onsdag den 13. januar 2021 kl. 21.52.36 UTC+1 skrev John Larkin:

some have UVLO on both high and low side voltage

Reply to
Lasse Langwadt Christensen

I don't think that helps the bootstrap problem.

It does help to pull the output node down to ground with some resistors or something... in some cases.

Reply to
John Larkin

onsdag den 13. januar 2021 kl. 23.18.20 UTC+1 skrev John Larkin:

it stops it from trying to turn on the high side when it can't and forces you to do the right things which is, charge the bootstrap cap before you start switching and keep the duty cycle in a range that keeps the bootstrap cap charged

Reply to
Lasse Langwadt Christensen

UVLO for both high and low side is essential for robust design

About the bootstrap, you always turn the low side on at startup to charge the cap. Some even apply small duty cycle to limit the inrush current

Common for bootstrapped half bridge is that 100% duty cycle is not allowed, at least not for many longer periods

--
Klaus Kragelund
Reply to
Klaus Kragelund

Right, but this approach is not intended for "long enough output pulses". Use an SN6501 and a tiny trafo wound on a 6-10mm toroidal core and there will be no low frequency limit.

Isn't it kind of obvious?

No high-performance driver I know of has this feature. SN6501 is dirt cheap and can power more than one driver.

Best regards, Piotr

Reply to
Piotr Wyderski

Am 13.01.21 um 21:52 schrieb John Larkin:

I had good results with these when biasing a GaASFET:

<
formatting link
>

A LED and some photo cells. Does not create any noise.

Gehard

Reply to
Gerhard Hoffmann

If the controller is counting on the upper driver to perform in response to a falling charge pump voltage in a predictable way, adding an external source could open a new can of worms.

I don't see an inhibit line . . . .

RL

Reply to
legg

You really want the low-side transistor on when possible to keep the bootstrap cap charged. And, then, the low-side needs to be turned on some of the time, when the PWM is off, to keep it charged.

Jon

Reply to
Jon Elson

One resistor is a lot easier than adding more logic.

--

John Larkin      Highland Technology, Inc 

The best designs are necessarily accidental.
Reply to
jlarkin

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.